Perceived obstacles in the undertaking of the
entrepreneurial career: an empirical study on
STEM and business alumni
Francesco Guberti
Dissertation written under the supervision of
Raffaele Conti and Alfonso Gambardella
Dissertation submitted in partial fulfilment of requirements for the MSc in International Management, at Universidade Católica Portuguesa and for the MSc in Economics and
Abstract page
Title: Perceived obstacles in the undertaking of the entrepreneurial career: an empirical study
on STEM and business alumni
Author: Francesco Guberti
Keywords: entrepreneurship, entrepreneurial propensity, perceived obstacles, alumni,
business, STEM, incubator, team diversity
English version
The goal of this study is to inquire the perceived expected and encountered obstacles in undertaking a career in entrepreneurship, both among university students and entrepreneurs. These perceptions were studied considering their academic background, with a major categorization between business and STEM alumni. The sample was drawn mainly among Italian citizens. Typical results have been found, such as a gender gap in the entrepreneurial propensity and a high proportion of entrepreneurs that had a mixed founding team composition. Despite no significant differences were found among university students, there are instead elements that distinguish and influence the perceptions among entrepreneurs. In particular, the experience in an incubator and the heterogeneity of the team composition appear to be perceived differently. The first element is evaluated in a relatively more favourable way by STEM alumni, while the same is true for the team composition by business alumni.
Portuguese version
O objetivo deste estudo é investigar a percepção dos obstáculos esperados e encontrados no início de uma carreira em empreendedorismo, tanto entre estudantes universitários como empreendedores. Estas percepções foram estudadas tendo em consideração os seus percursos acadêmicos, com uma subdivisão principal entre alumni de gestão e de CTEM. A amostra estatística foi retirada maioritariamente entre cidadãos italianos. Resultados espectáveis foram encontrados, como a disparidade de género na propensão a uma carreira de empreendedorismo e uma alta proporção de empreendedores que tiveram uma equipa fundadora com percursos acadêmicos e profissionais distintos. Embora não tenham sido encontradas diferenças significativas entre os estudantes universitários, existem elementos que distinguem e influenciam as percepções entre os empreendedores. Em particular, a experiência numa incubadora e a heterogeneidade na composição da equipa parecem ser percebidas de maneira diferente. O primeiro elemento é percebido de forma relativamente mais favorável pelos alumni de CTEM, enquanto o mesmo acontece com a heterogeneidade da composição da equipa pelos alumni de gestão.
List of contents
Introduction ... 4
Topic presentation ... 4
The two main definitions of entrepreneurship ... 4
New forms of entrepreneurship ... 5
The presentation of the subjects under analysis ... 6
Academic and managerial relevance ... 7
Problem statement ... 8
Research questions and expected findings ... 9
Literature review ... 12
The entrepreneurial intention... 12
The human capital success factors ... 13
Technical versus soft skills in the start-up environment ... 14
The demographic variables... 15
The role of networks and incubators ... 16
The obstacles towards entrepreneurship ... 16
Methodology and data description ... 17
Overall description of the first survey ... 18
Overall description of the second survey ... 19
Statistical analysis and empirical findings ... 21
The results from the first survey ... 21
The text analysis ... 21
The quantitative analysis ... 23
The results from the second survey ... 26
The comparison between business and STEM alumni entrepreneurs ... 26
The differences with respect to the mediating factors ... 30
The scenario analysis differences ... 32
The comparison between students and entrepreneurs ... 33
Conclusions and discussion ... 35
Potential implications on education and the development of entrepreneurship ... 37
The limitations of the thesis ... 37
The suggested further topics of research ... 39
References List ... 43 Books ... 43 Papers ... 43 Web resources ... 47 Exhibits ... 49 Appendices ... 64
Introduction
Topic presentation
The broad area of research of this thesis regards entrepreneurship, and more specifically the character of the entrepreneur and the entrepreneurial intention. This area of research has ancient roots, as, for example, one of the first definitions of entrepreneurship can be attributed to Richard Cantillon (circa 1730), who already emphasized the risk-taking behaviour that characterizes this career path. Centuries later, a pivotal shift in the field took place thanks to the work of Schumpeter and his famous theories of creative destruction (Schumpeter and Opie, 1934). Two of the most influential definitions of entrepreneurship will be reviewed briefly, in order to set the base for the subsequent discussion and in particular to highlight some fundamental characteristics that will be under study during the analysis.
The two main definitions of entrepreneurship
The milestone definition issued by Schumpeter sees the entrepreneur as an innovator who implements a change within one or more markets. This can be achieved through various means, such as the introduction of a new or improved good or service, or through an indirect innovation, for example affecting the methods of production (e.g. redefinition of the fundamental supply chain or sources of assets) or delivery. Moreover, the author highlights that “The entrepreneur–innovator’s motivation includes such aspects as the dream to found a private kingdom, the will to conquer and to succeed for the sake of success itself, and the joy of creating and getting things done.”, thus putting a strong emphasis on the motivations that bring a person to undertake this career path. Summarizing, this definition gives the opportunity to extract two main elements, the innovative input and the personal drive.
However, the definition remained subject to new and different denotations. Shapero (1975) had brought another important contribution through his definition: “[…] entrepreneur takes initiative, organizes some social and economic mechanisms, and accepts risks of failure”. Through this definition we can extract further layers of analysis in two fundamental elements. First, a social dimension, which encompasses the implications of the technical changes highlighted by Schumpeter on the people within the reach of those changes. Indeed, under any innovation there is an underlying social factor, which affects the innovation itself and gets affected by it. This element helps to remind the importance of the context in which an innovation occurs. Second, he talks about failure and risk. These elements address the
human side of entrepreneurship, by highlighting that, by launching a venture, a choice is made by the entrepreneur through a decision-making process influenced by various elements. The entrepreneur essentially weights the potential benefits and the potential losses.
In sum, the framework that can be extracted by the two definitions can be summarized as follows. The choice and process of launching a new venture has personal and external factors, or more accuratelly, both endogenous and exogenous variables that are present. On top of this differentiation, there is a further categorization possible, on technical and social elements. This is the framework on which the analysis of this thesis is based on.
New forms of entrepreneurship
Despite the great amount of literature on the topic, no single convention has been established on the definition of an entrepreneur. On the contrary, new related definitions have risen in the academic and professional worlds. One example, which is rather common nowadays, is the so called “intrapreneur”.
The definition of an intrapreneur is an individual that acts and behaves like an entrepreneur while working within an established organization. When looking at the above-mentioned definitions, an intrapreneur would fit them all to an almost complete extent. Thus, one could note that if the aim is to study the entrepreneurial intention, then also studying intrapreneurs would fit the purpose. However, intrapreneurs were not included within the scope of analysis as they indeed differentiate with respect to the entrepreneur, despite acknowledging that the boundaries are blurred. In fact, it is quite difficult to establish clearly what an “entrepreneurial” behaviour really means, let alone measure it. Thus, a direct comparison would be potentially misleading, as on one side we have a clear set of actions that signal an entrepreneurial intent, which follows the definitions cited above (e.g. risk-taking behaviour by capital commitments), while on the other we do not.
This inappropriateness of comparison is true especially due to one major difference between the two set of people: incentives. As argued by Scott Kirsner in his Harvard Business Review article (2018) they fundamentally differ, among other elements, with respect to the upside potential, the failure-related risks, the difference in environment in which they operate and the level of persistence. While these two latter characteristics’ fit may be subject to argument, the first two directly affect the incentive systems that influence their behaviour. Indeed, these elements have an extremely strong impact that cannot be disregarded and that
relate to the human side of the entrepreneurial activity.
Another relatively newly defined category of entrepreneur is the social entrepreneur, which is tied to the emergence of the social enterprises during the last years (Battilana and Lee, 2012). They differentiate with respect to the traditional definition of entrepreneurial venture by their mission, which has a strong and clearly stated social component, the source of financial sustainability, which includes both revenues and donations, and its role in society, being dually focused on social value and economic wealth generation, being a hybrid between a non-profit and a for-profit.
The presentation of the subjects under analysis
While acknowledging the difficulty in pinpointing a clear definition of an entrepreneur, especially due to the rise of so many facets in the field, in this research there will be a focus on market entrants, in accordance with a common convention. However, it is worth mentioning that one could argue that the character of the entrepreneur has not strictly set boundaries and that the elements that constitute it are potentially flexible, as for example nor age or size of the venture is an ultimately defining element (van Praag and Versloot, 2007). Moreover, I will focus on entrepreneurs that are engaged in innovative sectors or that propose a particularly innovative solution on the market. The choice of focusing on those that deal with a high level of innovativeness is due to the higher emphasis in these cases on the risk-taking behaviour. Indeed, their context brings to the extreme case the acceptance of a risk of failure. Either their conceived business models or products/services are new, most likely at least in part unprecedented and thus characterized by many risks and unknowns.
In integration to the perception of the obstacles related to a career in entrepreneurship, I will also cover some aspects related to some elements that could mitigate the limitations faced. I will specifically refer to incubators and the founders’ team composition, as major external influences.
While a great level of academic effort has been put to delineate the characteristics that commonly define an entrepreneur, I believe it is important also to study their perceptions taken from their own perspective rather than focusing only on observable characteristics. More in particular, the specific field of study concerning the behavioural determinants and motivations of entrepreneurs is a remarkably challenging and multifaceted. As detailed further below, I think that the entrepreneurial propensity is deeply influenced by one’s background, and thus I will focus my attention on this aspect.
Academic and managerial relevance
Many start-ups and new ventures had shown to have a strong impact on the economic activity, which can be extremely relevant and, in some cases, up to the point of redefining industry standards (Weinberger, 2018). Thus, studying the entrepreneurial intention becomes crucial to grasp the complex dynamics characterizing this process. Indeed, by defining a clear picture we may understand the causes of the rise and fall of new ventures, this latter phenomenon being widespread as noted in the paper “Why do Most Firms Die Young?” (Cressy, 2006). The rate of failures is indeed high, with examples of data showing 75% death rate in 10 years ("Entrepreneurship and the U.S. Economy", 2018).
This is particularly true in this period, characterized by high level of unemployment in a relatively significant amount of countries, especially regarding youth (see exhibit A). This represents a severe issue, to which entrepreneurship can be an alleviating factor (Birch, 1979). Another potential positive effect of better entrepreneurship is on developing countries (Kempner, 2017), by creating virtuous dynamics of new companies and employment generation. It is widely established that innovative entrepreneurs can bring great value to the whole economy (Solow, 1956), and at broader terms they play a key role in generating innovations (Acs and Audretsch, 2005).
The phenomenon of widespread and effective entrepreneurship can reach such a large scale that may affect the overall competitiveness, sustainability and strength of an entire national economy. The leading example of Israel can be taken as a reference. As described in the book “Start-up Nation: The Story of Israel's Economic Miracle” (Senor, Singer & Peres, 2011), despite the relative young age of the state, its small population and limited amount of raw resources and materials, the propulsion generated by the entrepreneurial drive within the population greatly enhanced its economy. Indeed, the famous NASDAQ index is populated by many tech companies founded in Israel ("NASDAQ - Non US companies", 2018).
Especially in these years, many waves of innovations and new business models are re-shaping and disrupting the economic landscape. It is such a vibrant period that there is even a definition to group some of these major changes: industry 4.0 (Marr, 2018). This allows entrepreneurs to exploit the potential offered by these changes, and leverage the economic momentum in order to establish themselves in the market (see Exhibit B to see the graph of Bloomberg’s U.S. start-ups barometer from 2007 until February 2018).
If there is so much potential, why then just a relatively tiny percentage of the population undertakes this type of career? Indeed, only 8.1% of the adult population in Europe is in the process of starting a business or has started a business in the last 42 months. Moreover, only roughly 29% of them are characterised by a focus on innovations ("GEM 2017 / 2018 global report", 2018).
By understanding what obstacles are perceived and encountered, we can identify potential biases and incorrect expectations that may hinder the undertaking of such a career ("Entrepreneurs anonymous", 2017). This may have profound implications in terms of education and knowledge diffusion. In fact, by diminishing the incorrect expectations of potential entrepreneurs, for example by providing relevant data and information, we can have a double beneficial effect. First, this change may allow improving the information available to the pool of potential entrepreneur candidates regarding the founding of a new venture. Hence, we may expect a better self-selection process, that is, the candidates who are the most fitting with respect to this particular career would improve their chance of ending up pursuing it, while on the other hand, those who are relatively less fitting would reduce theirs. This would improve the overall efficient allocation of human capital in the market with consequent potential positive economic outcomes. Second, those who choose this career would be already informed of the most critical aspects to be aware of. This could potentially avoid costly mistakes or incorrect focuses.
Problem statement
The entrepreneurial intention has often been studied in various terms, especially on attempts on profiling a typical entrepreneur (more details in the literature review section). I will try to focus particularly on the academic background of entrepreneurs, and see if it influences the expectations regarding the obstacles to be faced in becoming entrepreneurs, and if there are differences also with respect to the perception on the obstacles met. Much emphasis has been placed on trying to pinpoint certain characteristics that define a “typical entrepreneur” and that can predict a career in entrepreneurship. My focus will be different, as I will directly enquire the entrepreneurs’ perceptions, and not limit the research on visible characteristics and behaviours.
More in detail, I will focus on two categories of people, those with a business background (a definition that includes economics, management, finance, marketing etc.) and those with a STEM background (which comprehends engineers, physicists, chemists etc.).
Indeed, these two broad categories of subjects are the academic areas from which most entrepreneurs come, and are thus to be considered as representative of the population of entrepreneurs. For example, two thirds of innovative entrepreneurs in Italy had a business background while the remainder a technical one ("Ecosistema Italia 2015: l'innovazione in un’infografica | StartupItalia!", 2015).
The entrepreneurs’ academic path may instil a specific framework, point of view, approach and organization of priorities. This difference may have implications on both the entrepreneurial intention and after the launch of a new venture. The link between education and entrepreneurial intent has already been highlighted in the literature (Gavron, 1998), (Audretsch, 2012).
Research questions and expected findings
The research question can be defined as such:
Are there differences between business and STEM alumni in the perception of both expected (pre-launch of the start-up) and encountered obstacles (post-launch of the start-up), in undertaking the entrepreneurial career in innovative sectors?
My main hypothesis is that there are, as business and STEM alumni start with different assumptions and expectations regarding two fundamental elements, at an exogenous and endogenous level. First, both their priorities and their perceived personal endowment differ, in particular with respect to what is needed in terms of skills and resources for a successful start-up. Second, their assumptions regarding the level and the severity of the impact by external elements on the creation and development of the venture are different.
More in detail, I believe that these two set of people will regard differently two elements that are fundamental to launch an innovative venture, the soft skills and the technical aspects. Indeed, each set is more versed and educated than the other in one of the two elements. I expect that their respective lack of knowledge will make the two parties keener in indicating different aspects related to the founding of a venture as more hazardous and difficult to overcome. This means that, given a certain academic background, people experience the undertaking of a career in entrepreneurship differently.
I will now list the differences that I expect to find from the analysis. STEM alumni overlook relatively more the “soft” side of starting a venture. Their technical focus (“hard skills”) will make them focus less on the relational dimensions, such as the importance of
having a strong network or the team building process, or the emphasis on communicating effectively their idea. They may know better how to estimate the technical feasibility of a product or process, but may not estimate as well as their counterparts the economic ways of exploitation. This may lead to overconfidence if they perceive their product/service as technically superior, while neglecting the “soft skills” side.
Business students instead are on average less capable of estimating technical risks and costs, while being more prone on focusing on the soft skills and relational side. This because education path leads them to do so and they are relatively more experienced in dealing with such matters. For example, business courses often entail some presentations and most business schools offer communication and negotiation courses.
This potential overarching and idiosyncratic difference, derived from their background, will also be tested and checked for confirmations in this study through the analysis of students’ answers, in a separate survey. The problem statement can be broken down into three main hypotheses:
H1 - Business and STEM alumni perceive different expected obstacles pre-foundation (i.e. before the launch of the venture) on 9 main dimensions (that are listed in the methodology and data description section, such as building a strong team, obtaining funding etc). That is, before the launch of their start-ups, due to their difference in background, they differ in the expected level of potential impact derived from a set of adverse issues, both endogenous and exogenous.
H2 - Business and STEM alumni perceive differently encountered obstacles post-foundation (i.e. after the launch of the venture) on the same 9 main dimensions mentioned in H1. That is, once they have launched their ventures, due to their difference in background, they perceive different levels of impact derived from a set of adverse issues, both endogenous and exogenous.
H3 - Business and STEM alumni perceive a different level of mitigation with respect to the issues presented, derived by either team composition or activities in an incubator. That is, both categories perceive and recognise a dissimilar level of contribution that fills the gap that they have in terms of skills and experiences. This contribution is granted, when applicable, either by their colleagues/co-founders that have a different background or by the activities undertaken within an incubator.
The first two hypotheses allow to understand whether there are indeed consistent differences in perceived expected as well as encountered obstacles (respectively pre and post the launch of the start-up), which are derived from the difference in academic background. The third hypothesis will try to uncover whether these differences are perceived as partly mediated, for both categories, by external influences such as other co-founders with different backgrounds or by experiences in incubators. Indeed, these elements could provide different points of view and build complementary skills with respect to those already possessed by the target entrepreneur. The results of this study have highlighted that H1 has been disconfirmed, as students present a solid homogeneous set of answers. On the other hand, H2 and in particular H3 have been confirmed at a statistically significant level. The differences appear to be most radical before the launch of the venture, and there is a strong evidence regarding the varying effects that the mediating factors have on the types of entrepreneurs.
It is worth noticing that there are some entrepreneurs having a mixed academic background. They have been accounted for, as they provide valuable insights as special cases. Indeed, having a hybrid academic background, they may be endowed with the best range of skills and points of view to tackle effectively a career in entrepreneurship. Indeed, differences were also registered through their answers.
Literature review
I will present the main topics that influence the entrepreneurial behaviour and its perception and that were incorporated in the empirical analysis.
The entrepreneurial intention
The character of the entrepreneur is multifaceted and complex in nature, since entrepreneurship is a multifaceted phenomenon and not a single defined business function. Indeed, there is no single fitting profile and nobody is born as entrepreneur, being a process rather than a stand-alone quality. There are however certain dimensions that affect this process and influence the entrepreneurs’ heterogeneous activities, that happen at different levels (such as idea formation, funding gathering et cetera). The most widespread traits and cognitive feature that are propaedeutic to become an entrepreneur are related to the willingness to achieve, the risk and ambiguity tolerance, an internal locus of control, the perception of self-efficacy and the keenness on putting objectives (Shane, Locke & Collins, 2003). Researches focused on the codified “Big Five” behavioural predispositions (Costa & McCrae 1985), trying to link it to venture survival rate. Conscientiousness was find significant, while surprisingly a negative relationship was found with the openness predisposition (Ciavarella, Buchholtz, Riordan, Gatewood & Stokes, 2004), and the remainder were found as not correlated.
A psychological dimension that influenced much the studies on entrepreneurship regards the self-efficacy. Despite being susceptible to changes during the experience, it has been regarded as an important factor to consider. Although no clear direct relationship with venture success is firmly recognized (Baum & Locke, 2004), the research on its fundamental roots may bring more clear conclusions on self-efficacy’s role in the entrepreneurial process.
More directly in terms of entrepreneurial intention, there are still gaps to be filled (Fayolle & Liñán, 2014). However, this research field has been widely scrutinized in order to find potential predictors for venture initiations (Krueger & Carsrud, 1993) based on attitudes, behaviour and the influence of the social environment. Regarding this latter component, cross-countries studies are particularly useful to understand the impact of the social stigma related to failure on the entrepreneurial intentions (Autio, H. Keeley, Klofsten, G. C. Parker & Hay, 2001). A remarkable effort, in terms of gauging the entrepreneurial intentions, have been conducted by the Panel Study of Entrepreneurial Dynamics (PSED - The University of Michigan", 2018), which collects data, among other types, related to the motivations to launch
a venture. Further studies tried to narrow down the potential reasons for the choice of such a career path to fewer options, to get an understanding of the most relevant trends, and in this particularly cited paper there are six: “self-realization, financial success, roles, innovation, recognition, and independence” (Carter, Gartner, Shaver & Gatewood, 2003). Interestingly, despite the diversity of the ventures founded, one of the consistently mentioned reason along the financial one is the social recognition (Cassar, 2007). In terms of education, both the least and most educated set of people have higher rates of entrepreneurship (Blanchflower, 2000).
Another important element to be considered is the opportunity identification. Entrepreneurs, especially in innovative sectors, must be able to individuate and assess an opportunity to enter effectively into the market. However, despite being so intuitively important, it has been rather difficult to study it in a comprehensive way (Gaglio & Katz, 2001). As a further complication, the next step in the process, the process of the undertaking, has also recently being subject to debate (Palich & Ray Bagby, 1995). In terms of risk-taking for example, studies suggest that rather than more risk-takers, entrepreneurs categorize ambiguous business scenarios more positively. It has emerged that much of this process has to do with frameworks and the personal assessment of opportunities and threats.
While it has been established that entrepreneurial education affect the entrepreneurial orientation, little is known of the exact contextual dynamics, especially when business and STEM students are confronted (Maresch et al., 2016), (Zhang, Duysters and Cloodt, 2013).
The human capital success factors
The next step is to inquire who has success among those that launch a venture. More in detail, the intent is once again to pinpoint the characteristics and the trends that could help individuate the most widespread success factors. In this sense, there are endogenous and exogenous forces that might shape the eventual outcome. Among the endogenous, i.e. those elements that are embedded in the entrepreneur, there are both innate (e.g. gender) and developed characteristics (e.g. education). The exogenous ones, instead, are derived by the context and environments in which the entrepreneur operates.
One of the most influential elements among the endogenous variables is education. Interestingly, through a literature review a trend was noticed, in which education had no impact on entry but it has an impact with respect to performance (Van Der Sluis, Van Praag & Vijverberg, 2008). Another element is the presence of an entrepreneurial background in the family (Dunn & Holtz-Eakin, 2000). Indeed, there is a higher tendency from those who have
members in their family to be entrepreneur themselves. This affects through their financial possibilities and, more importantly, through their experiences and results.
Regarding the finances of entrepreneurs specifically, the heterogeneity in cases, contexts and situations make it difficult to have a clear trend (Hurst and Lusardi 2004). However, the relationships seem to remain flat for most of the distribution, while phenomenon like liquidity constraints may still be present and elements such as inheritance may have an alleviating outcome (Holtz-Eakin, Joulfaian & Rosen, 1994). One of the most impactful element in the early stages is the prior entrepreneurial experience (Stuart & Abetti, 1990).
Technical versus soft skills in the start-up environment
As seen in the introduction, the entry of innovative firms has a positive effect in terms of innovations, especially in sectors near the technology frontiers (Aghion, Blundell, Griffith, Howitt & Prantl, 2009). The venture’s human capital is essential to its success, regarding both the founding members and the first employees (Rocha, Vera & van Praag, Mirjam C. & Folta, Timothy B. & Carneiro, Anabela, 2016).
One of the few papers that studies the impact of technical and soft skills in early stage ventures has the research purpose to see if there is complementarity between them (Mueller & Murmann, 2016). The sample is drawn from German start-ups, with data derived from the German Federal Employment Agency. Their results show how teams with mixed skills are more likely to introduce innovations in the market. Interestingly, their findings apply only when the founder has technical skills and then hires business alumni. Indeed, this result does not even apply when the two sets of skills belong to the founder nor when the founding team is mixed. This is explained by the diminished costs related to deciding a joint strategy. Moreover, another suggestion regards the imbalances in terms of knowledge, since entrepreneurs with STEM background are more business savvy with respect to how much business alumni are tech experts. In addition, they find that founders with a technical background do not often hire business alumni.
The balance of skills has been reviewed and it is widely established that entrepreneurs must be jack-of-all-trades, or in other word, must be able to face different challenges effectively (Lazear, 2004). To do so, it requires having several skills, which may be grouped into categories, for example in soft and technical skills as it is the case in this study.
The demographic variables
The most relevant element among the demographic variables is gender. In fact, there is a substantial gender gap that persists between countries regarding entrepreneurship. The effects are multifaceted, as fewer ventures are launched by women (6 for every 10 male entrepreneurs, "GEM Global Entrepreneurship Monitor", 2018), and they are usually characterized by lower levels of performance (Jennings & Brush, 2013). This phenomenon has its roots in the self-perception that is widespread in many cultural contexts, regarding skills, capabilities and role of the woman in society. This is fuelled also by the imaginary that links entrepreneurship to a prevalently male activity (Gupta, Turban, Wasti & Sikdar, 2009). Moreover, the gender gap in the STEM subjects also has an indirect effect on the entrepreneurial intention of women. While there seems to be no bias in terms of debt financing (Buttner & Rosen, 1989), there seems to be a negative effect on equity financing.
Age’s effect on entrepreneurship is still controversial ("Entrepreneurs Get Better with Age", 2013). The widespread convention is that each generation has its own idiosyncrasies with respect to entrepreneurship, combined to the effect brought by the stage of life in which each generation is at a given time. Millennials in particular possess some characteristics, such as above average IT skills and relative better education, that would lead to strong and sustained entrepreneurial activities ("State of Entrepreneurship: Mixed Indicators Prompt Call for Entrepreneurial Renewal", 2015). Moreover, there is empirical evidence of the relatively easier capacity to adapt to new information (Parker, 2006). However, factors such as debt burdens and adverse economic conditions (e.g. the financial crisis of 2008) negatively influenced the rate of start-up foundations. This generation has nevertheless not entered the peak age (40 years) and it is expected then to see an increase in its rate. On the other hand, older generations can rely on broader experience, larger networks and more consistent financial resource (Weber & Schaper, 2004). It is highlighted in the GEM report that entrepreneurs aged from 25 to 44 are the most active entrepreneurs. Moreover, in the U.S. tech companies are founded by middle-aged entrepreneurs, despite the most famous cases involving the foundations of Microsoft and Apple at a relatively early age (Wadhwa, Freeman & Rissing, 2008). In addition, in this latter study, it is noticeable that the majority held bachelor degrees (92%), while 31% held a master degree and only 10% hold a PhD. Half of these were in STEM subjects and one third in business, which mirrors the situation in Italy, both in terms of education and other demographics. Returning to the age dimension, there are quite a few pieces of evidence that indicate that start-ups founded by middle age entrepreneurs achieve
faster growth (Henley, 2005), (Brüderl & Preisendörfer, 2000).
For the sake of brevity, matters regarding the background of entrepreneurs in terms of ethnicity and culture are not covered in detail. Nevertheless, it is worth mentioning that cultural aspects have profound impact on the entrepreneurial orientation and the socio-economic context in which new ventures operate (e.g. immigration forces that are highly correlated with entrepreneurial activities) (Lee & Peterson, 2000) (Mueller & Thomas, 2001).
The role of networks and incubators
In terms of networks, it has been established that the early stage dynamics are fundamental to the firms’ performance (Eisenhardt & Schoonhoven, 1990). In particular, the team composition and its set of experiences has an impact in terms of venture strategy, as for example to focus on exploitation or exploration (Beckman, 2006). Indeed, the myth of the lone entrepreneur has been abandoned, while no clear relationship between team heterogeneity and performance has been found (Klotz, Hmieleski, Bradley & Busenitz, 2013).
Incubators and accelerators contribute to the development of new ventures. In the world there are widespread and established organizations carrying these types of activities ("Seed-DB | List of individual Seed Accelerator programs", 2018) ("Organizations | Crunchbase", 2018). The role and impact of incubators has been long studied. It is interesting to notice that it has been mentioned that the degree of technical focus of the product/service of the venture has an influence in terms of outcomes for start-ups that go through an incubator (Cooper, 1985). However, it is difficult to pinpoint the activities of an incubator, due to the wide range of processes and specifics that are possible (Aernoudt, 2004).
The obstacles towards entrepreneurship
Regarding the obstacles, some datasets on the topic gather these sorts of data, however in a very heterogeneous way ("The 9th Annual State of Entrepreneurship Address", 2018). Few studies highlighted the sources of fear and failure of new ventures (Cacciotti, Hayton, Mitchell & Giazitzoglu, 2016) (Cacciotti, Hayton, Mitchell & Allen, 2015). As noted in this latter paper, fears and obstacles can both inhibit and motivate entrepreneurs. The authors’ efforts are towards unifying the framework regarding the fears and obstacles faced by entrepreneurs, and will be used as a guideline to define the basis for the empirical research.
Methodology and data description
The analysis has been carried out only through primary data collected by means of two main surveys (please refer to appendices 1 and 2), which will be used for both descriptive and inferential purposes. The choice of the survey method has been driven by the necessity of tackling the variety in the samples under analysis. Indeed, while an interview might have provided with more in-depth insights, it lacks the standardised format of information that provides for direct comparability, which is essential for the purposes of this study. The information collected will be both qualitative and quantitative. These latter will be gathered through 7-points Likert scales. The advantage of this process is that I gathered data that precisely fit my research question, and this is a fundamental feature since I am also inquiring very narrow and specific personal perceptions that have not yet been studied in-depth. Both endogenous and exogenous variables were considered, as well as hard and soft skills related ones. The survey is an effective method since what it is looked for is the perception of the person, and thus the matter of subjectivity is not a bias, but instead exactly what I want to study and it is therefore rightly incorporated. One comparable example is provided by a similar study in terms of methodology and relative scope regarding perceptions: “Kitchen confidential? Norms for the use of transferred knowledge in gourmet cuisine” (Di Stefano, King & Verona, 2013).
These perceptions have been gathered in two manners. First, through an open question, that allowed respondents to express themselves without any bias or guidance derived by the proposed closed items. It has been included because it addresses specifically the lack of in-depth and freedom of expression that lie beyond the closed items which are typically asked in a survey. Thus, this kind of question has always been asked at the very beginning of the respective surveys. Secondly, the remaining sections comprised closed items that allowed for a greater level of comparability between respondents on defined topics.
However, it is to acknowledge that the potential sample will be limited using this approach, if compared to analysing secondary data provided by established organizations. All the surveys’ data were gathered with total anonymity of the respondents. This feature contributes to a high level of honesty and avoidance of biases with respect to the answers provided.
The surveys have been handed out on a web format through the well-known Qualtrics platform, and the data gathered has been subsequently analysed using IBM’s SPSS statistical software. The respondents have been personally contacted through means of social networks,
namely Facebook and LinkedIn, given the characteristics showed through their profile (e.g. type of employment). They have been informed of the aim and scope of the survey and its modalities, and then provided with the relevant link in case of agreement to contribute with their opinion. This personal contact and approach has ensured that the respondents were exactly the persons having the required background which is within the scope of the analysis, and that only those people were provided with the access to their respective surveys.
Regarding the entrepreneurs, a first set was found and targeted through incubators platforms and websites. Namely, founders of start-ups operating in the Start-up Lisboa and Start-up Braga incubators were contacted (see the section regarding the web resources for the relevant links). These incubators have been chosen for their national reach and established recognition and to permit cross-culture validations’ type of analysis. Moreover, they represented an optimal choice since they host start-ups with a heterogeneous variety of backgrounds and fields of operation. Indeed, by focusing on more narrowly focused incubators, for example hosting start-ups developing only maritime technologies and solutions, there would have been a potential bias and distorted representation of the overall start-up scene. The second set was provided by researching entrepreneurs that were graduated from two of the most prestigious universities in Italy, which are the Politecnico di Milano and Bocconi University.
Regarding the data collection from students, the survey has been shared through the Facebook social media, by contacting those students belonging to closed Facebook groups of their respective courses, either official (see the section regarding the web resources for some examples of relevant links) in their nature or not. Now that the methodology of distribution of the surveys has been covered, the next section will highlight their contents.
Overall description of the first survey
The first survey has been distributed among current university students of both business and engineering university programs (namely from the following universities: Bocconi, Católica Lisbon School of Business & Economics and Polytechnic University of Milan), ranging from the second year of bachelor up to the last year of master. This survey had three main goals. Firstly, to analyse and understand the main obstacles students perceive with respect to the entrepreneurial career and their overall entrepreneurial orientation. Secondly, to see whether business and STEM students’ perceptions present consistent differences, again also in terms of their entrepreneurial intention. Finally, this survey also had the function to adjust the
questions to be asked to current or recent ex-entrepreneurs, which is the target sample of the second survey, in an exploratory approach.
Overall description of the second survey
The second survey will measure the very same perception on the obstacles related to entrepreneurship proposed in the first survey, but this time with the addition of a temporal dimension. More in detail, the second questionnaire asks about the entrepreneurs’ expected severity regarding a certain set of issues both before the launch of the venture, and the level of severity of these same when they were encountered, i.e. post launch. By doing this, it is possible to rank the perceived obstacles and put them into macro-categories of severity.
In this way, the entrepreneurs (who, by definition, have shown a strong entrepreneurial tendency by choosing this peculiar career option), disclose both what they expected to face and what they actually encountered along their experience. This allows also to understand how much of their expectations were correct, in addition to whether there is a consistent difference among the two sets of alumni in terms of expected and encountered obstacles. Moreover, the survey asks to evaluate, when applicable, the perceived impact on skills mediation (which refers to H3 described above) derived by the influence of either colleagues with different backgrounds or by the influence of an experience in an incubator program. Lastly, demographics have also been collected, to deepen the overall analysis and provide a holistic understanding of the sample under study. For this category segment, a wider range in terms of age was expected, since becoming entrepreneur is traditionally less age-bounded relatively to undergoing higher education.
It is worth noticing that I decided not to include in the analysis also early joiners (i.e. non-founders) for two reasons. First, because they have different incentives in terms of personal economic gain and achievement. Indeed, founding a venture has deep implications in terms of commitment and economic risks. Nevertheless, one could argue that early joiners could negotiate a deal that includes a stake in the company. However, there is a second reason that led me to exclude this category, that is, they would not experience the same problems and issues with respect to the founders. They most likely would join a company at least partly defined, in which time and economic resources have been already expended, and where many adjustments and solutions have been already applied. Some issues may not even be present anymore (e.g. the founding and bootstrapping money being already secured). This strong element of difference does not allow making appropriate comparisons with respect to the
perceived obstacles, especially when the analysis touches the perceptions at the pre-launch phase.
For a detailed breakdown and analysis of each item of the two surveys, please refer to appendix 3. Please refer to appendix 4 for the description of the complementary tools used in the analysis.
Regarding the choice of statistical analysis, since the objective was to find evidence of quantitative differences within the responses given some categorical variables (i.e. background and type of activity presently carried out), the ANOVA set of techniques was defined as the most suitable. In particular, the ANCOVA technique was chosen in order to conduct the analysis, given its feature of being able to take into account covariates that may have an influence on the relationship between the dependent and independent variables. This helps to reduce the error variance, eliminate the disturbance of confounding variables and thus achieve an overall lower level of bias. On the other hand, inserting these controls also diminishes the risk of suffering from an omitted variable bias. In the analysis and exhibit sections, only statistically significant and most important results are shown for the sake of conciseness. Moreover, all ANCOVA analysis were manipulated as to obtain the highest level of adjusted 𝑹𝟐 possible, excluding all the non-relevant covariates, while taking care that the parameter of the main independent variable is not significantly impacted. A common statistical threshold of p=0.05 was adopted to evaluate the magnitude of the statistical significance. Regarding the assumptions, the dependent variables are always measured on a continuous scale. The independent variables are two or more categorical and independent groups and the observations are independent one on the other. Since the scales are bounded there is no outlier that could bias the analysis. Moreover, residuals are normally distributed for each category of the independent variable. In addition, the covariates are assumed to be linearly related to the dependent variable at each level of the independent variable. Homogeneity of regression slopes are also assumed. Scatterplots of residuals were analysed to check for any abnormal behaviour. Homogeneity of variance is tested through the Levene’s test as well as homoscedasticity that is tested through the scatterplot of expected and observed standardized residuals. Other linear regressions were used to support the analysis, such as in the case of the scenario analysis. Regarding the qualitative responses, the above-mentioned algorithm was used to conduct a text analysis.
Statistical analysis and empirical findings
In this section the information gathered from the surveys is reviewed, and their elaboration to draw the most relevant outcomes, which will be the basis for the following discussion section. Thus, I will describe also the formal empirical procedures and statistical analysis used to draw these conclusions. Moreover, I will highlight the differences following the comparison tool for the analysis cited in the methodology section, i.e., formulating cross category comparisons between all the proposed category subdivisions.
The results from the first survey
The number of full respondents to the first survey has been of 126 students. Details in terms of the demographics can be found at Exhibit 2.
The text analysis
A first useful step is to evaluate through descriptive statistics the most evident findings. In particular, the qualitative section can already grant us some useful hints to guide the analysis. As mentioned in the methodology section, to study this set of answers, a text analysis algorithm has been used on the open question submitted to students. The analysis allowed to understand what were the most cited obstacles by STEM and business alumni. This has been determined by considering the relative frequency of citation (expressed in percentage) with respect to the total number of respondents per category of background. Some evident trends emerged, which will be now described. Please refer to exhibit 3.
The first interesting element is to notice that, between students having different backgrounds, there is some sort of homogeneous and coherent evaluation. Indeed, in relative (percentage) terms the frequency of obstacles cited do not vary apart from just three exceptions. For example, for both categories of students, the funding obstacle was cited the most (46% of frequency for business and economics students, 42% for engineering students). This is by far the clearly most felt obstacle, as it detaches itself by 10% points in both categories from the respective second most cited obstacles. Then, a second set of obstacles clearly stands out for both categories of students. The first obstacle comprised in this set regards the external economic environment (30% for business and economics students, 25% for engineering). This comprehends the market uncertainties, the presence of competitors, the changing customer preferences and all the exogenous elements that determine a negative landscape to launch a
venture. Related to this obstacle, a second one comprehended in the same set is the potential loss of personal economic stability and other personal constraints, such as the hardship in forming a family while managing a start-up. The third obstacle is the risk aversion behaviour (28% for business students, 22% for engineering students). This comprehends the potential lack of courage or belief in one’s own skills and capabilities or even a sheer lack of courage. It is interesting to notice that two of the top cited obstacles are strictly related to the personal aspect of launching a venture, instead of being related to the technical aspects.
A third set of obstacles (whose percentage is stable above 10%) comprises three obstacles. The first one is the bureaucracy involved. This encompasses the pressure exerted by the necessity to comply with complex legal environments, fill the amount of bureaucracy, red tape etc. The second one relates to the lack of adequate education or experience to launch a successful venture (10% for business students, 17% for engineering students). The third obstacle is related to the hardship in finding the right idea. This encompasses the level of novelty of the idea, its potential for scaling, or even just the chance of finding one which is worth pursuing (18% for business students, 14% for engineering ones).
The last set of obstacles comprises all the minor remaining ones, which do not reach a frequency of 10% in the answers of both students’ categories. The first obstacle being the tax pressure and fiscal burden (9% for business students, 11% for engineering students). This is interesting to notice, as start-ups usually benefit from fiscal incentives and advantages (see the web resources section for an example referring to the Italian region of Lombardy, "Assolombarda - Guida alle agevolazioni per le Startup innovative", 2018). However, it is worth noticing that this might be due to the very specific cultural contexts in which the respondents live: both Italy and Portugal have high level of taxes. The second obstacle of this set regards the difficulty in finding the founding team’s members and possessing a strong and diverse network of professionals to rely on (11% for business students, 3% for engineering ones). The last one is related to the concept of culture (4% for business students, 1% for engineering students). Students described that one obstacle is overcoming the general culture which lies in the context in which they live. This means for example the approval of their family on the matter, the difficulties in breaking a more traditional career path or not feeling to be having the appropriate cultural framework to tackle such a complex endeavour.
students, it is also interesting to observe that there are three major differences. The first and largest difference is about the team and network obstacle. Engineering students perceive it as much less relevant (a difference of 8%) than their counterparts. This most likely indicates a lower perceived need to complement their skills through a diverse team composition. The second largest difference is about the education and experience necessary to launch a start-up (7% of difference). In this case, engineering students signalled that this obstacle is perceived as relatively more burdensome. Lastly, also in the case of the loss of personal economic stability we see engineering students giving more weight to this obstacle (6% difference). All other differences were equal or lower than 5%, and were thus deemed as not relevant.
This qualitative part allows to set the benchmark for the subsequent analysis. Indeed, the main points found, i.e. the general level of perception among the obstacles, the general homogeneity of perception among the two student categories and their few distinct differences, will be checked through the quantitative analysis.
The quantitative analysis
The quantitative part is divided into two main set of questions. The first one gives an indication of the entrepreneurial orientation. The second one has the aim to confirm the perceived severity of obstacles. Please refer to exhibit 4 for the table of the values considered in the analysis for the first set and exhibit 5 for the second one.
The first set sees again a strong level of homogeneity between the two categories of respondents. The first question, regarding the level of consideration that respondents have had with respect to a career in entrepreneurship, shows an average of 4.43 (7 being the most extreme positive answer and 1 the most negative one, which applies to also the other two questions), meaning a slightly positive response overall. Engineering’s higher average and lower standard deviation signal that they indeed had taken on average more in consideration this career choice (0.18 more than their counterparts). On a similar trend, this is also confirmed by their answers to the third question, which asks whether they would prefer to be entrepreneurs or employees. Here again the average is quite high (4.94 with 1.78 of standard deviation) and engineers signal a higher preference with a slightly lower standard deviation. This means that overall, this type of career prospect is seen quite positively by both categories of students and it is felt as preferable to the standard condition of employment. However, the second question, regarding
the level of information gathered on pursuing this career, shows a considerably overall lower result. Indeed, the general average is of 3.37, which means a slightly negative response. In this case, students of business showed a greater average value (3.44, distancing by 0.28 the value of the engineering students of 3.17). This lower value was expected since it represents a further step towards the pursue of this career choice. It is worth noticing that the higher value given by business students may be partly explained by the potential related content studied in some specific courses dealing with entrepreneurship.
The second set of questions allows us to define the relative severity of certain predefined obstacles. As in the former case, the most evident finding is the homogeneity of the responses among the two categories of students. Indeed, in terms of their ranking regarding the perceived severity of the nine obstacles, there is no difference. We can summarize the levels with three groups of obstacles. The first group includes the two obstacles with the highest average, of which the first one is related to the obtainment of funds (average of 5.64) and the second one to the risk of losing personal economic stability (average of 5.56), both with a relatively low standard deviation with respect to the answers provided by the two groups. The second groups of questions is formed by three obstacles, namely the “The capacity and chance of spotting an opportunity and estimating the potential economic value of the venture”, “Estimating the technical feasibility of the product/service” and “Building a team and having a professional network”. Their averages all fall within a narrow range (4.67, 4.72 and 4.77 respectively). Regarding the standard deviations, they are generally stable and homogeneous except the one related to the estimation of the technical feasibility for the engineering students (value of 1.78) while on the contrary it is very low for the business students (value of 1.39). This may be due to the relatively heterogeneity of major of the respondents, where some are potentially more inclined in educating hard knowledge that may be relatively more easily applied to evaluate innovative products or services (for example, a student of software engineering may be endowed with skills that better apply to the estimation of ideas that may become the basis for the launch of a start-up, rather than a civil engineer).
The third set of obstacles comprises all the remaining four, which are the least felt, namely the “The social stigma related to failure”, “Potential personal constraints (e.g. expected effort and time dedication requested higher than average)”, “The nicheness of the entrepreneurial career and the availability of other career options”, “The perception of one's own abilities and traits (among which, for example, the risk-taking behaviour) as not matching
those of a potential entrepreneur”. Their averages fall between the values of 3.74 and 4.31, respectively 3.77, 4.01, 3.74, 4.31. About their standard deviations, there are only two notable cases. First, the one related to the potential personal constraints for engineering students (value of 1.83). Secondly, the one related to the social stigma related to a potential failure (value of 1.75). Indeed, these are extremely personal obstacles and it could be expected to see more variability in terms of responses.
We can now examine more closely the differences between the two categories of students. There are three major values that vary in a considerable way. Namely, these are: “The capacity and chance of spotting an opportunity and estimating the potential economic value of the venture (and the potential related rewards)”, “The risk of losing personal economic stability”, “The nicheness of the entrepreneurial career and the availability of other career options”. All of them show higher values for engineering students, meaning that they perceive them as more impactful issues. The first obstacle shows a 0.38 difference, the second a 0.27 difference and the third one a 0.33 difference.
ANCOVA analysis have been conducted to find evidence of statistically significant differences among the averages of the above cited obstacles, when considering the student category as main independent variable and all the other demographic covariates as controls.
First, a factor has been created by considering the three questions regarding the entrepreneurial orientation. The extraction method chosen is the Principal Component one. As expected, only one factor had been extracted and therefore no rotation was needed. Please refer to Exhibit 6 to 11 for the details. An exploratory ANOVA was run on this factor, considering only education background and gender. Only this latter resulted statistically significant, with a lower average for female respondents (exhibit 12). Then, ANCOVAs were run on the three obstacles that presented the widest differences but none resulted statistically significant. The only obstacle that was almost significant (sig. =.079) was the one related to the capacity to spot and estimate the economic value of an idea (exhibit 13). However, its 𝑹𝟐 was of 3.3%, implying
that even if there is systematicity, it would account for little of the variation. This was confirmed also by the results of the ANOVAs that considered only the education background as independent variable. Please refer to exhibit 14.
The results from the second survey
Regarding the second survey, the number of respondents has been of 112 entrepreneurs. For a table of the demographics please refer to exhibit 15 and 16.
The subsequent part is divided into three main set of questions. The first one gives an indication of the perceived level of severity of predefined obstacles. The second one has the aim to describe the impact of mediating factors over the obstacles. The third one describes the scenario analysis. Please refer to exhibit 17 and 18 for the tables of the values considered in the analysis for the first set and exhibit 19 for the second one.
The comparison between business and STEM alumni entrepreneurs
As in the case of the analysis of the students’ answers, it is interesting to notice that there is a great homogeneity in the responses across the three categories of alumni. The obstacles, in this case, can be divided into three main sets.
The first one, which is formed only by the obstacle related to the obtainment of funds, since it is clearly standing out as the most severely perceived, both before and after the launch of the venture. Indeed, it reaches a value of 4.83 before and 5.23 after, with a high standard deviation in the value attributed “before” of 1.91. This latter value means that the expectations before were quite heterogeneous, and this appears plausible, since getting funds is a rather uncertain process. On the other hand, the difference among the above-cited numbers means that, not only it was already thought as the most severe issue to be confronted, but nevertheless they underestimated it.
Then a second set, which comprises most of the obstacles, include all those that were deemed as intermedium in their level of impact. These are namely: the capacity and chance of spotting and evaluating the venture, the estimation of the technical feasibility, the process of building a team and having a strong professional network, dealing with the government, the risk of losing personal economic stability, the potential personal constraints, and the market uncertainty. Their values range for the before category between 4.04 and 4.4, while for the after category between 4.14 and 4.85. The standard deviations are stable except for those regarding the variables about dealing with the government and the risk of losing personal economic
stability, especially before the launch of the start-up. Especially for this latter obstacle, it is expected some sort of heterogeneity in the responses, as the outcome could strongly depend on one’s own family and financial wealth background condition. The difference in the values between before and after do not appear relevant except for two cases. The first one relates to the obstacle concerning the process of forming a team and a strong professional network. Indeed, the difference in values amounts to 0.51 (from 4.34 to 4.85), meaning that also this obstacle has been generally underestimated by the respondents. The second one relates to the obstacle concerning the process of dealing with the government, where the difference amounts to 0.47 (from 4.33 to 4.80). Here again, this obstacle seems to have been underestimated to a significant extent.
The third set of obstacles comprises only three, which were deemed as the least negatively impactful. These are namely: “The perception of one's own abilities and traits (for example, the risk-taking behaviour) as not matching those of an entrepreneur”, “Lack of adequate academic preparation to begin a start-up” and “The social stigma related to a potential failure”. Their values remain below 4 both before and after, more specifically remain in the range between 3.07 and 3.52 for the before category, and between 2.82 and 3.52 for the after one. Their standard deviations remain particularly high across both before and after categories, and this seems reasonable, as they all are very personal elements. Interestingly, the obstacle related to the lack of adequate academic preparation to launch a start-up not only has the lowest value, but it’s the only obstacle which decreases if compared to the after category. In other words, their perception was that they underestimated the adequacy of their academic preparation to establish a start-up. Moreover, it is the only case of obstacle that was overestimated.
We can now examine more closely, by breaking down the answers, the differences between the three sets of entrepreneurs, both before and after. This allows for multiple level of analysis and comparisons. First, it is interesting to notice that the average level of severity expressed by business alumni before is 4.01 and 4.41 after, which means an overall underestimation. On the contrary, STEM alumni start with a slightly more negative expectation, with an expressed average before of 4.11, which proves to be worse than reality, as the average value of after is 3.96. Those with a mixed background follow a more similar path to their business-only counterparts, starting with a value of 3.99 and ending with a value of 4.17.
Secondly, we can break down the intra category differences, to reveal the presence of potential patterns. The standard deviations follow the general tendencies described above in the aggregate analysis, and are highly stable across categories.
Looking only at entrepreneurs with a business background, we can observe four obstacles that are characterized by a relevant change (with a magnitude bigger than 0.5) in value between before and after. These are the obstacles related to the estimation of the technical feasibility, the obtainment of funds, the process of dealing with the government and the risk of losing personal economic stability. All these variables are characterized by an underestimation, since their values after show an increase respectively of 0.53, 0.67, 1.07, 0.56. In particular, the difference regarding the process of dealing with the government shows an increase of strong magnitude. Overall, the average difference is of underestimation by 0.40.
Looking only at entrepreneurs with a STEM background, we observe instead also a different pattern. Indeed, out of the three obstacles that show a difference between the before and after values, two are overestimations. These are the obstacles related to the estimation of the technical feasibility and the capacity to spot an opportunity and evaluating its potential economic value. These show a decrease of respectively 0.62 and 0.60. On the contrary, the process of building a team and a professional network is underestimated by 0.60. It is interesting to notice that the average difference is of -0.15, which means an overall overestimation of the severity of the obstacles. Indeed, out of the 11 obstacles, 8 present negative values, or in other words were overestimated.
Taking into consideration only the entrepreneurs with a mixed background, we can observe four obstacles that show a relevant variation. These are the capacity to spot an opportunity and evaluating its potential economic value, the team building process and the creation of the professional network, the risk of losing personal economic stability and the potential personal constraints. Their respective differences are of 0.93, 1.07, -0.60 and 0.60. Only the third mentioned obstacle showed an overestimation while all the others, like in the case of business alumni, were all underestimations. All of these differences were confirmed through ANOVA analysis, with values around 10%.
At this point, it is also possible to analyse the differences between the three categories of alumni. Comparing the business and STEM alumni, we can highlight that most of the relevant
